National studies show that drug use rates have increased in the last decade among women, comprising a major public health concern in the US. However, women are greatly underrepresented in neuroimaging studies, and the paucity of studies that explicitly target sex comparisons in addicted populations contributes to a gap in the study of the sex specific neurobiological mechanisms underlying drug addiction. Over the last decade, in a series of magnetic resonance imaging (MRI) studies (conducted with previous support including R01DA023579, R01DA020949), we have thoroughly mapped the clinical symptoms of cocaine addiction to the neural networks underlying impairments in Response Inhibition and Salience Attribution (iRISA). This model proposes that the drug assumes heightened salience at the expense of non-drug related reinforcement as associated with abnormalities in reward processing and concomitant decreases in inhibitory control, together increasing addiction severity (including craving, a proxy of relapse) in susceptible individuals. The iRISA model highlights the role of the dopaminergically innervated prefrontal cortex (PFC) and its connections to mesolimbic and striatal subcortical regions as assessed functionally and structurally. However, the majority of this neuroimaging research has been accomplished in male individuals with cocaine use disorders (iCUD). In the current project we aim to expand the reach of iRISA by comparing equal numbers of male to female iCUD; to test this model?s generalizability (vs. drug specificity effects), we will also include individuals with opioid use disorder (iOUD). We will conduct functional MRI during reward processing, inhibitory control and cue-reactivity tasks, and, to inspect generalizability of results beyond task-related activations, during resting-state. Beyond functional activations and connectivity, anatomical scans will assess the underlying gray matter integrity. Across all aims, healthy controls will be included to establish norms. We hypothesize female iCUD to differ from male iCUD, or female controls, in a pattern indicative of enhanced vulnerability to iRISA inclusive of compensatory PFC activations and abnormalities in structural measures; iCUD vs. iOUD comparisons will be exploratory. The novelty of this proposal is further enhanced by an exploratory aim to compare, in a within- subjects design, menstrual cycle (and hormonal) effects and by developing sophisticated machine-learning algorithms to incorporate data from all imaging modalities to yield an automated group classification and addiction severity (including craving) prediction tool. Considering that the majority of research in addiction occurs in males, clarification of the sex differences in the neural underpinnings of iRISA could reinforce the importance of studying both genders and suggest that different treatment strategies may be effective in women (potentially of most impact when timed vis--vis menstrual cycle), contributing to the development of tailored (gender-based) treatment options. Including equal numbers of women and men would advance basic studies of drug addiction and ultimately save resources by minimizing cost and adverse effects in future clinical trials.